Bug 45052
| Summary: | Change TransformationMatrix to use floats rather than doubles | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Chris Marrin <cmarrin> |
| Component: | Platform | Assignee: | Nobody <webkit-unassigned> |
| Status: | RESOLVED WONTFIX | ||
| Severity: | Normal | CC: | ahmad.saleem792, dino, d-r, graouts, mattwoodrow |
| Priority: | P2 | ||
| Version: | 528+ (Nightly build) | ||
| Hardware: | PC | ||
| OS: | OS X 10.5 | ||
Chris Marrin
Changing the TransformationMatrix class to floats would reduce the precision of calculations. But it would match FloatPoint and FloatPoint3D, would correspond to the matrices passed to WebGL, and would lend itself much better to SSE2 optimizations, which can process single precision twice as fast as double precision.
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Ahmad Saleem
@Matt & @Antoine - do we have any such ambitions or we can close this?
Antoine Quint
I don't think so. Closing.